Some quick thoughts on opt out

In general, I have not opined much on the subject of “opt out,” for a number of reasons. First, there’s little/no good data or research on the topic, so my opinions can’t be as informed as I would typically like them to be. Second, I don’t know that I have much to add on the issue (and yet I’m about to give my two cents). Third, it’s a trend that actively worries me as someone who believes research clearly shows that tests and accountability have been beneficial overall. I don’t really see much policymakers can do to stop this trend short of requiring public school students to test [1].

Despite my best efforts to avoid the subject, over on Twitter, former MCPS Superintendent Joshua Starr asked me what I think of this EdWeek commentary on opt out. Here are some excerpts of their argument and my reactions.

First, the title is “Test-taking ‘compliance’ does not ensure equity.” Probably the authors did not write this title, but it’s a very weak straw man. I know of few/any folks who believe that test-taking compliance ensures equity. I certainly don’t believe that. I do believe having good data can help equity, but it certainly doesn’t ensure it.

Some parents have elected to opt their children out of the annual tests as a message of protest, signaling that a test score is not enough to ensure excellence and equity in the education of their children. Parents, they insist, have a right to demand an enriched curriculum that includes the arts, civics, and lab sciences, and high-quality schools in their neighborhoods.

I don’t have good evidence on this (I don’t think anyone yet does, but hopefully several savvy doctoral students are work on this topic), but my very strong sense is that the folks opting out of tests are not typically doing it as an equity protest. Everything I’ve seen and heard so far says this is largely, but not exclusively, a white, upper-middle class, suburban/rural phenomenon [EDITED TO ADD: Matt Chingos has done a preliminary analysis of this issue and largely agrees with this characterization: My conversations with educators in California, for instance, suggest that the high rates of opt-out in high schools in some affluent areas are because the exam was seen as meaningless and interfering with students’ abilities to prepare for other exams that actually matter to students (e.g., APs, SAT).

Since it was signed into law in 2002, No Child Left Behind has done little to advance the educational interests of our most disadvantaged students. What’s more, the high-stakes-testing climate that NCLB created has also been connected to increased discipline rates for students of color and students with disabilities.

I think the first sentence there is not correct–as I showed in the previous post, there’s evidence that achievement has increased due to NCLB for all groups, including the most disadvantaged (but not much evidence it has narrowed gaps). I’m not aware of well designed research showing the latter claim, but that’s not my area. Regardless, as I also discussed in the last post, sweeping claims of harm to disadvantaged students are hard to square with empirical evidence on outcomes such as test scores and graduation rates.

And even after these tests reveal large outcome gaps, schools serving poor children of color remain underfunded and are more likely to be labeled failing. Most states have done nothing to intervene effectively in these schools, even when state officials have taken over school districts. Moreover, despite NCLB’s stated goal of closing the achievement gap, wide disparities in academic outcomes persist.

I think this is mostly true, though of course it depends on state (some states are much more adequate and equitable in their funding than others). And the lack of intervention in low-performing schools really is about a lack of effective intervention, though I’d be very curious what interventions these authors would recommend. It’s true that achievement gaps persist, though I believe racial (but not income) gaps are about as small now as they’ve ever been.

We are not opposed to assessments, especially when they are used for diagnostic purposes to support learning. But the data produced by annual standardized tests are typically not made available to teachers until after the school year is over, thereby making it impossible to use the information to respond to student needs.

Some of the new state tests get data back faster. For instance, some California results were made available to teachers before the end of the year. In general I think it’s a bad idea to heap too many different goals for a single test. It’s not clear to me that we always want our accountability test to also be our formative, immediate feedback test–those probably should be different tests. But that doesn’t necessarily obviate the need for an external accountability test.

Thus, students of color are susceptible to all of the negative effects of the annual assessments, without any of the positive supports to address the learning gaps. When testing is used merely to measure and document inequities in outcomes, without providing necessary supports, parents have a right to demand more.

Again I think the intention of both the original NCLB and the waivers was that, in the early years of school “failure” students would be provided with additional supports and options (e.g., through supplemental education services and public school choice) to improve. Those turn out not to have worked, and perhaps future supports will not either, but it’s not necessarily for lack of effort. I’m curious what specific supports these authors would advocate, bearing in mind the intense hostility among half our nation to raising any additional funds for schools or anything else.

The civil rights movement has never supported compliance with unjust laws and policies. Rather, it has always worked to challenge them and support the courageous actions of those willing to resist. As young people and their allies protest throughout the country against police brutality, demanding that “black lives matter,” we are reminded that the struggle for justice often forces us to hold governments and public officials accountable to reject the status quo. Today’s status quo in education is annual assessments that provide no true path toward equity or excellence.

This strikes me as a stretch, though I agree with the first half of it. I’m not sure the “black lives matter” movement was really about holding the government accountable to reject the status quo, as much as it was about holding both government and individuals accountable for centuries of unjust laws and actions (but this is not remotely my area).

The anti-testing movement will not be intimidated, nor is it going away.

I think that’s right. Though reducing or eliminating teacher accountability based on state tests would probably at least reduce the extent to which the unions are actively encouraging opt-outs.

Some may choose to force districts to adopt a more comprehensive “dashboard” accountability system with multiple measures. Others may push districts to engage in biennial or grade-span testing, and still others may choose to opt out. What remains clear is that parents want more than tests to assess their children’s academic standing and, as a result, are choosing to opt out of an unjust, ineffective policy.

With respect to the first sentence, some states did this (and all states had the opportunity to do this in their waivers). With respect to the sentence, it’s not clear to me how biennial or grade-span testing is any more “just” than yearly testing. Perhaps if these authors stated what they think is the optimal testing regimen from a “justice” perspective, that would help.

So, I don’t think it’s an especially convincing argument. But I don’t know that the pro-opt-out movement really needs convincing arguments. If parents have the right to opt their kids out of tests, at least some of them will do so. I suspect this will lead to increased inequity, but that’s an empirical question for another day.

[1] Were I omnipotent, I would enact that rule, and I’d also require private and homeschool kids to test.


8 thoughts on “Some quick thoughts on opt out

  1. To me, “opt out” should be a parental right. Whatever their reasons, they have the right to make decisions for their child. As educators, our job is to help them make an informed decision — based on ACCURATE information and not hype. However, what’s lost in the argument is that participation rate requirements were meant to end the practice of schools encouraging low performing students to be absent on testing days. That should absolutely continue to be discouraged. But we need to distinguish between subversive behavior of school personnel and a parent exercising his/her civil rights. Otherwise, we will never win this debate.


    • I guess I’d say that, if it’s a parent’s right to opt their kids out of testing, it’s a parent’s right to opt their kids of anything that happens in schools. In other words, I very much worry about the precedent.


      • But don’t we already have a precedent? You can opt your child out of sex ed. You can opt your child out of a field trip. You can request (demand) a change in teachers. You can opt to home school. We might debate the wisdom of some of the choices parents make, but I hope we can defend their right to make them. Where we draw the line is when the choice of one parent affects another student. That’s where it gets blurry to me. Can we make the argument that one parent opting to remove his/her child from testing can ultimate harm another student?


  2. I think the strongest equity argument against testing is that it adds a hugely time-consuming, stressful, expensive, and demoralizing facet to low-SES kids’ school experiences, without yet delivering much change in return.

    You are quite right to point out that much of the country doesn’t want to increase funding, which definitely limits options for change. No wonder parents want to take some or all of the money, time, and teacher energy currently oriented toward tests and redirect it back to the arts, or science, or PE, or other school programs. This is absolutely the main motivation I’ve heard for opting out in my family’s district (which has a quite high opt-out rate). If teachers were also being evaluated largely on test scores here, that would probably be an even stronger motivator for opting out.

    My children have taken standardized tests for years, and as a teacher I think SBAC may well be an improvement over the previous state tests, at least for older children. Nevertheless, our experience this year was probably the last straw for me regarding testing for my younger son. You can read it here, along with the reservations I’ve developed about SBAC testing as a teacher:

    I don’t normally post anonymously, but since the story involves his privacy as well as mine, it seemed like a good idea.


    • Thanks for the personal comment and story. It sounds like in your son’s case, opting out was definitely the right move. As for the money, time, and teacher energy devoted to tests, some thoughts …
      1) It’s really not much money–in the scheme of educational spending, it’s a percent or so.
      2) The actual testing time is not much. Again a percent or so. Now, the fact that districts have said “I’ll see your state tests and raise you four benchmarks” is a district choice, and I think an unfortunate one. Were I a parent, I’d be absolutely pushing my district to cool it with the benchmarks and test prep and just let the state tests be a natural and small part of the school year. Perhaps easier said than done, but it’s the district responses to the tests that seem like the problem, not the tests themselves.
      3) With respect to teacher energy, pretty much the same thought applies. I strongly suspect that the teaching that will get the best bang for the buck is not teaching to the test, but general good teaching. Focus on teaching kids well and they’ll do well on the tests.


  3. “Everything I’ve seen and heard so far says this is largely, but not exclusively, a white, upper-middle class, suburban/rural phenomenon.” This isn’t really true in Oregon. Portland, the biggest city, has a high opt-out rate (9% overall). It’s definitely higher for white and multiracial students than for other racial/ethnic groups, but the lowest rate, for Asian students, is still pretty high at 4% (compared to 10% for multiracial and 11% for white students). As a researcher, you’ll probably love seeing actual data: . It’s fair to say the schools with the highest opt-out rates are generally more middle class/upper middle class than the schools with low opt-out rates, but it’s not a direct relationship.

    “The actual testing time is not much. Again a percent or so […] cool it with the benchmarks and test prep and just let the state tests be a natural and small part of the school year.” I teach math. I did no test prep this year, but it took 7 school days to administer the SBAC out of our 178 day year (which really isn’t 178 days of math; we lost at least 5 days to field trips, end of school activities, etc.). That’s about 4% of my time with those students… and of course some had to have make-up or extra days of testing, too. And I regretted not doing test prep; in theory, I agree completely with you that their skills should be tested adequately without the prep, but the computer interface has things some of them are not familiar with (like * for multiplication).

    I also have some reservations about the SBAC generally (though honestly for math it seems likely to be an improvement over the multiple-choice-only former state test, even in its first year). I think the tests may be traumatic for some of the younger kids, especially. And I found some (though not all) of the critiques of the Math SBAC here fairly chilling: It doesn’t surprise me a test would be less than perfect in its first year, but it’s annoying that there isn’t more public acknowledgment (especially from the feds) that we shouldn’t expect too much from it right away.

    Having said all that, I do basically support the idea of testing, generally. I think it’s important to be able to compare schools and even teachers, and to act to fix glaring problems. The degree of precision policy-makers seem to expect to get terrifies me, though. I haven’t been a teacher for long, but I can see that the standardized test data from my students has been way too noisy to give a fair picture of how I’m doing: some years I’d look great and others lousy, even though my teaching was about the same. Among other factors, someone has to test in March and someone gets to test in late May, because there aren’t many computers… that’s a pretty big chunk of the year’s standards that the March teacher’s kids didn’t learn before taking the test. If a principal wants to get rid of a teacher, s/he’ll surely have him or her testing in March.

    Also, it really offends me how secretive the testing companies are about the tests, and how restrictive they are about what we teachers can say about them. If, hypothetically, I had been called over by a student because the buggy computer interface wouldn’t let him enter the correct answer to 1 of his 38 problems, and if, hypothetically, I privately agreed it was impossible, neither my student nor I would have had any way to report the error, and in fact I wouldn’t even be able to talk about it. Hypothetically.

    Thanks for explaining your thinking so clearly in your columns here and for keeping an open mind. It’s been interesting reading your point of view. As a teacher who supports the Common Core, I hope we can find a resolution that helps improve our schools. Right now I think that at a minimum, parents need a more convincing explanation of what the heck we’re doing with testing, and I appreciate that you’re trying to provide part of that explanation.


      • Interesting, and I suspect it might actually be fairly consistent with Portland (which is whiter and richer than its suburbs).

        I know it’s not fair to expect a single analysis to answer everything, but these are some questions it raises for me:

        How hard is it to opt out in NY? Does it vary by district or school? Does it require forms? meetings? I strongly suspect administrators are more deferential to parents’ wishes when the parents are upper middle class (especially if they’re lawyers!), so schools with more of those families might be more accommodating about opt-outs. Higher-SES parents’ work schedules may also be more flexible for meetings, and I believe they’re more likely to have two-parent homes. And I’d guess upper middle class parents are far more likely to be aware of the option to opt out and how to do it, because they have more time to read all this stuff and are more likely to be native English speakers.

        The analysis concludes, “[D]istricts with lower test scores have higher opt-out rates after taking socioeconomic status into account. Potential explanation for this pattern include district administrators encouraging opt-outs in order to cover up poor performance, districts focusing on non-tested subjects to satisfy parents who care less about standardized tests, and parents becoming more skeptical of the value of tests when their children do not score well.” I teach at an arts focus school, so the second reason stands out for me. But his third seems skewed; as Chingos said in the previous paragraph, “[J]ust because lower-scoring districts have higher opt-out rates (controlling for free/reduced lunch) does not mean that lower-scoring students are more likely to opt out. It could be the higher-scoring students in those districts that are doing the opting out.” That’s what I hear on the grapevine: parents resent the school being judged by math and reading standardized test scores, so they pull their high-scoring kids to make a point and stop being (as they see it) complicit in a broken rating system.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s